MolSoft LLC

Introducing GINGER from MolSoft: a cutting-edge software designed for lightning-fast high quality conformer chemical library generation on GPUs. State-of-the-art generative Neural Network-based algorithm to efficiently explore the conformational space of molecules, providing diverse, low-energy conformers. Read more... https://www.molsoft.com/GINGER.html

If you missed MolSoft’s “Advances in Physics and AI Methods for Drug Discovery” meeting we have uploaded recordings of some of the talks online here at our YouTube Site https://www.youtube.com//videos . A huge thank you to all the exceptional speakers who shared their insights and to everyone who attended. Your engagement made this event a resounding success!

MolSoft Molecules in Silico - YouTube MolSoft ICM Help Videos

Join us for an exciting free online event: Advances in Physics and AI Methods for Drug Discovery - MolSoft ICM UGM 2023. Features talks about the latest cutting-edge technologies such as GPU-based and AI-accelerated virtual screening, structure- and ligand-based drug design, PROTAC modeling, membrane protein modeling, and much more. Click the link to register https://us06web.zoom.us/webinar/register/WN_iLQ-iFX5QSubkSCdDWcq6Q #/registration and see the schedule here www.molsoft.com/training.html

We want to let you know about V-SYNTHES - a new Virtual Screening approach available in ICM or provided as a service by MolSoft.
https://www.nature.com/articles/s41586-021-04220-9
MolSoft’s ICM software is the docking, screening and cheminformatics engine that powers the recently published structure-based V-SYNTHES virtual screening approach developed by a team of researchers from University of Southern California, ChemSpace, Enamine and ten other organizations. V-SYNTHES application to compound docking screening led by Prof. Katritch, first reported in January 2022 in Nature, iteratively narrows down a subset of compounds for structure-based docking. The approach reported experimental hit-rates that were twice as high as regular screening and identified new highly potent hits with broad diversity and new scaffolds for a kinase (ROCK1) and GPCR target (CB1/2).

The approach relies on highly efficient MolSoft's Biased Probability Monte Carlo method, optimized force field, to dock and score smaller chemical probes to regions of the drug target pocket. The fragments do not have to be tested experimentally but instead high scoring ones are enumerated into the full chemical from the giga-size library (Enamine REAL Space – 11 billion chemicals) using ICM-Chemist enumeration tools. The enumerated library is then docked and scored again using Molsoft-ICM-VLS. This hierarchical protocol accelerates the docking and scoring run and makes it applicable to very large databases. The flexibility of the binding pocket was represented using MolSoft's 4D docking methodology which uses an ensemble of structures efficiently for fast and accurate flexible-receptor ligand docking.

The ICM scripts for running V-SYNTHES have been made available on GitHub. MolSoft can also perform the screening as a service and deliver the results in approximately 2-3 weeks. To run this pipeline outside Molsoft, V-SYNTHES license, and ICM-Pro-backend + VLS Graphics-front-end licenses are required. An efficient setup would benefit from over 200 ICM Cluster/Cloud backend licenses to parallelize the jobs.

#

Synthon-based ligand discovery in virtual libraries of over 11 billion compounds - Nature V-SYNTHES, a scalable and computationally cost-effective synthon-based approach to compound screening, identified compounds with a high affinity for CB2 and CB1 in a hierarchical structure-based screen of more than 11 billion compounds.

New publication from MolSoft. A novel bond graph-convolutional NN
architecture for learning statistical torsion profiles (GCNN-
STP). Application for quick qualitative strain calculations in CADD and conformational sampling.
https://pubs.acs.org/doi/10.1021/acs.jcim.2c00790

MolSoft is excited to announce the release of a new ICM-Pro version 3.9-3. The new version contains a Torsion Profile Neural Network (NN) Prediciton Engine, Protonation State of Ligand vs pH tool, Compact Efficient GIGA Sized Screening Libraries, Automated PROTAC Modeling, Molecular Dynamics (MD), New NN Ligand Docking and Screening Score and much more...
https://www.molsoft.com/new-features.html

Molsoft L.L.C.: New Features RIDE is a fast 3D molecular similarity search method based on Atomic Property Fields, developed at MolSoft. RIDE searches databases of compound conformers for molecules that are isosteric to the query, i.e. have similar 3D configurations and distributions of atomic properties.